[Battlemesh] Battlemesh v5 tests

[Antonio Quartulli]

Hi Clauz,

I would like to propose another test, a bit different from the others.

I'd like to see how batman-adv (and the other protocols if the feature is
supported) behaves in case of a client roaming from one to another mesh node.

Actually I worked on this feature for batman-adv during last year, but I've
never practised a medium/large scale test. It would be nice to see how
connections are kept alive while moving (e.g. an ssh session, a skype call).

Would it be feasible?

Cheers,

On Sat, Mar 10, 2012 at 02:57:50PM +0100, Martin Hundebøll wrote:
> Hi Clauz,
> 
> I would like to introduce another test for throughput. I work with the network coding[1], and would like to see some tests that show some scenarios where network coding is feasible and some where it is not.
> 
> For example, the chain topology is a nice scenario:
> 
> 	A -- B -- C -- D -- E
> 
> because network coding can increase throughput at each hop, if data goes in both direction. However, when data goes in one direction only, our implementation delays packets at each hop.
> 
> Also, two different paths on the same channel:
> 
> 	A -- B -- C
> 
> 	D -- E -- F
> 
> because network coding can reduce the air traffic and thus increase the total throughput (again, when data flows in both directions.
> 
> What do you think?
> 
> Kind regards
> Martin Hundebøll
> Wireless Communications
> Aalborg University
> 
> +45 61 65 54 61
> mhu at es.aau.dk
> 
> [1] http://www.open-mesh.org/wiki/open-mesh/2011-08-18-network-coding-first-steps
> 
> On 03/01/2012 01:35 PM, Clauz wrote:
> > Hi!
> > As some of you know, last December, during the battlemesh meeting that
> > we held in Berlin, I was asked to take care of the tests to be performed
> > on the mesh that we will be deploying in Athens.
> >
> > So, here are some ideas:
> > http://www.battlemesh.org/BattleMeshV5/Tests
> > Feel absolutely free to criticize, discuss, propose and give feedback :)
> >
> > ciao,
> > Clauz
> >
> > _______________________________________________
> > Battlemesh mailing list
> > Battlemesh at ml.ninux.org
> > http://ml.ninux.org/mailman/listinfo/battlemesh
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh

-- 
Antonio Quartulli

..each of us alone is worth nothing..
Ernesto "Che" Guevara
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
URL: <http://ml.ninux.org/pipermail/battlemesh/attachments/20120311/7747d047/attachment.pgp>